Aesthetics of neural network robots

Issues arising from ‘Neurotic’

Neurotic and the aesthetics of music

with 2 comments

This blog has now moved to http://neuroticpogoingrobots.wordpress.com/

But, basically it is dedicated to the aesthetic and scientific issues that arise from Neurotic, where punk bands perform to 3 giant robots that have been educated in classic punk (fuller description below).

Neurotic clearly references the validity of formalism, enhanced formalism, the emotive properties of music (should they actually exist) and the hardy perennial issue of whether or not humans are just complex machines. Arguments about a distinction between machine and humans that are based on man’s appreciation of ‘beauty’ are clearly under scrutiny here: Neurotic’s robots are ‘controlled’ by neural network programmes designed to mimic the brain’s mirror neurons. Mirror neurons, so it is claimed, govern our empathetic response to the world. So, do the robots empathize with the music? Do they enjoy the music?

In the robots we also see the possibility of a socially value free enjoyment of music – these robots have no heirs and graces, no normative social beliefs that would prohibit the enjoyment of any form of music. They don’t have cultural baggage – is their judgment objective as a result?

Join the debate about whether Neurotic’s robots are developing taste as their preferences develop.

For more information see www.fiddian.com or www.andrewtweedie.de.

Andrew Tweedie (Director of Music, Neurotic)

 
Advertisement

Written by andrewtweedie

19 May, 2008 at 3:05 pm

Posted in Uncategorized

2 Responses

Subscribe to comments with RSS.

  1. Please join in the debate!

    andrewtweedie

    19 May, 2008 at 3:18 pm

  2. Hi,

    I need to know more about how you ‘trained’ the robots (unless there is an explanation somewhere that I have not been able to find yet, if so please point me to it…). When you say they were ‘exposed’ to punk records, what exactly do you mean? What musical elements actually inform this learning? Is it the ‘events’ of the music, the structure (repetitions etc), is it the loudness, the rhythms, etc., etc.? Do the robots develop a probablility distribution of certain pitches etc according to the data they are trained on? Are they trained on one set of stimuli to the excclusion of others (for example, some models of listener perception are trained only with regards to pitch, and some apply ‘multiple viewpoints’ in order to model perception on many different musical elements and the relationship between them)?

    I would be really interested to know the details!

    Vanessa

    12 June, 2008 at 2:19 pm


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Connecting to %s

Follow

Get every new post delivered to your Inbox.